Power Consumption, Temperature, Noise, Overclockability
We already measured the power consumption of the reference Radeon HD 5870 in our earlier tests, but the ASUS Matrix 5870 is an original solution. So, we measured its power draw on our specially designed testbed configured like follows:
- Intel Core 2 Quad Q6600 CPU (3GHz, 1333 MHz FSB x 9, LGA775)
- DFI LANParty UT ICFX3200-T2R/G mainboard (ATI CrossFire Xpress 3200 chipset)
- PC2-1066 SDRAM (2x2 GB, 1066 MHz)
- Enermax Liberty ELT620AWT PSU (620 W)
- Microsoft Windows 7 Ultimate 7 64-bit
- CyberLink PowerDVD 9 Ultra/"Serenity" BD (1080p VC-1, 20 Mbit)
- Crysis Warhead
- OCCT Perestroika 3.1.0
The new testbed for measuring electric characteristics of graphics cards uses a card designed by one of our engineers, Oleg Artamonov, and described in his article called PC Power Consumption: How Many Watts Do We Need?. As usual, we used the following benchmarks to load the graphics accelerators:
- CyberLink PowerDVD 9: FullScreen, hardware acceleration enabled
- Crysis Warhead: 1600x1200, FSAA 4x, DirectX 10/Enthusiast, "frost" map
- OCCT Perestroika GPU: 1600x1200, FullScreen, Shader Complexity 8
Except for the maximum load simulation with OCCT, we measured power consumption in each mode for 60 seconds. We limit the run time of OCCT: GPU to 10 seconds to avoid overloading the graphics card's power circuitry. Here are the obtained results:
We did not expect the Matrix 5870 to be an economical solution as it is targeted at users who don’t care much about that factor. Its power consumption in every mode is comparable to that of the dual-processor Radeon HD 5970 rather than of the single-processor Radeon HD 5870. The card’s first 8-pin power connector is loaded twice as heavily as the second connector and its peak current may be as high as 14 amperes, which equals about 170 watts. We wouldn’t recommend saving on your power supply for this card. A high-quality 600W or better PSU should be used together with the Matrix 5870.
Using the direct-touch technology and a fast fan, the cooling system of the Matrix 5870 copes with its job but doesn’t show anything special. It is even inferior to the reference cooler from AMD in the 3D mode. The ASUS card’s higher heat dissipation should be taken into account, yet we think that this cooler might do even better if it were not for some flaws in its design that we described in the previous section of our review.
The Matrix 5870 is not good in terms of noisiness considering that the background noise in our test lab is 37 dBA. Our testbed itself is not silent, even with the new CPU cooler and power supply, but the Matrix 5870 was perfectly audible when we launched heavy 3D applications. A reference Radeon HD 5870 is much quieter even under load. Alas, this is another consequence of the flaws in the cooler’s design. For example, the Gigabyte GV-R587SO-1GD, even though not exactly silent, did not irritate us with the rattle of its blower as the ASUS card does.
As for overclocking, we used iTracker 2 to increase the GPU and VDDCI voltages to 1.4 and 1.226 volts, respectively, and make the GPU of our Matrix 5870 stable at a frequency of 1010 MHz or 1 GHz.
The graphics card passed all our tests at that frequency but produced much more noise. The graphics memory could only be overclocked to 1225 (4900) MHz. We couldn’t reach the rated frequency of the K4G10325FE-HC04 chips, which is 1250 (5000) MHz. This may have been due to the cooling system because it doesn’t cool all of the GDDR5 chips installed on the card. Anyway, our overclocking experiment was successful enough for us to be interested in its practical benefits. Therefore we will benchmark the ASUS card at the overclocked frequencies as well.